- Title
- Beyond curve fitting? Comment on Liu, Mayer-Kress, and Newell (2003)
- Creator
- Heathcote, A.; Brown, Scott
- Relation
- Journal of Motor Behavior Vol. 36, Issue 2, p. 225-232
- Publisher Link
- http://dx.doi.org/10.3200/JMBR.36.2.225-232
- Publisher
- Heldref Publications
- Resource Type
- journal article
- Date
- 2004
- Description
- Y.-T. Liu, G. Mayer-Kress. and K. M. Newell (2003) fit learning curves to movement time data and suggested 2 new methods for analyzing learning. They claimed that the methods go "beyond curve fitting.' However, in neither their curve fitting nor their new methods is measurement noise accounted for, and therefore they produce inefficient and biased results. Using the data of Liu et al., in which variance caused by learning is small relative to the level of noise for most participants, the present authors demonstrate those problems and provide better alternatives that are more noise tolerant, more powerful, and go beyond curve fitting without displaying the extreme bias produced by the methods of Liu et al.
- Subject
- averaging; learning curve; model testing; motor learning
- Identifier
- uon:2019
- Identifier
- http://hdl.handle.net/1959.13/27931
- Identifier
- ISSN:0022-2895
- Rights
- This is an electronic version of an article published in The Journal of Motor Behavior Vol. 36, Issue 2, p. 225-232. The Journal of Motor Behavior is available online at: http://www.tandfonline.com/openurl?genre=article&issn=0022-2895&volume=36&issue=2&spage=225
- Language
- eng
- Full Text
- Reviewed
- Hits: 888
- Visitors: 1475
- Downloads: 309
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | SOURCE1 | Author final version | 625 KB | Adobe Acrobat PDF | View Details Download |